<html>
  <head>
    <meta content="text/html; charset=ISO-8859-1"
      http-equiv="Content-Type">
  </head>
  <body bgcolor="#FFFFFF" text="#000000">
    <font face="Helvetica, Arial, sans-serif">Hi Pedro,<br>
      <br>
      Sorry for confusing the thread. I was responding to Nat's point
      about structured headers. It's not really relevant to the issue
      you are addressing. As for requirements for the back-channel call,
      you can add a document on the wiki, or file a new issue on the
      site as an enhancement for the working group to address and then
      put in the ticket all the current requirements. Others can then
      comment on the ticket and the working group can track it.<br>
      <br>
      Note, that for this back-channel capability to be relevant to an
      RP, the RP must support the concept of "server side" sessions (or
      maintain a "black list" of revoked sessions). This doesn't tend to
      be capabilities that most RPs support.<br>
      <br>
      Thanks,<br>
      George<br>
      <br>
    </font>
    <div class="moz-cite-prefix">On 3/13/14 11:12 AM, Pedro Felix wrote:<br>
    </div>
    <blockquote
cite="mid:CAD+AFDsa8qPeBv41_eGGJzPqoAte7cCyxcPyV9-C-GBM3cw8Kw@mail.gmail.com"
      type="cite">
      <div dir="ltr">1) Since I'm rather new in this group, what would
        be the best way to continue this discussion? In this email
        thread? By trying to produce a requirements doc on the wiki?
        <div>Most probably, I will be working on an implementation of
          this feature in the near future.<br>
          <div><br>
          </div>
          <div>2) Picking up on Justin's reply: an approach would be to
            also use the "aud" and the "sub" to identify the session to
            cleanup. I don't like the idea of requiring a round-trip to
            the introspection endpoint in order to check the token
            purpose. Makes sense?</div>
        </div>
        <div><br>
        </div>
        <div>Thanks</div>
        <div>Pedro</div>
        <div class="gmail_extra"><br>
          <div class="gmail_quote">On Thu, Mar 13, 2014 at 2:12 PM,
            Justin Richer <span dir="ltr"><<a moz-do-not-send="true"
                href="mailto:jricher@mitre.org" target="_blank">jricher@mitre.org</a>></span>
            wrote:<br>
            <blockquote class="gmail_quote" style="margin:0 0 0
              .8ex;border-left:1px #ccc solid;padding-left:1ex">
              <div bgcolor="#FFFFFF" text="#000000"> A number of our
                apps use this combined approach -- the server sends out
                a signed JWT that the client can check the "iss" field
                and signature, then (if it cares to do so) the client
                does introspection with the server at "iss" to see if
                the token's still valid and what it's good for.<span
                  class="HOEnZb"><font color="#888888"><br>
                    <br>
                     -- Justin</font></span>
                <div>
                  <div class="h5"><br>
                    <br>
                    <div>On 03/13/2014 09:48 AM, George Fletcher wrote:<br>
                    </div>
                    <blockquote type="cite"> <font face="Helvetica,
                        Arial, sans-serif">On the "structured token"
                        side of things I remember having a discussion
                        about this at IIW (a few back) and I thought
                        someone and written something up. It was needed
                        in a number of cases that were using the token
                        introspection endpoint as a way to identifier
                        the authorization server to send the token to
                        for introspection. I can't find my notes on the
                        conversation but maybe someone else remembers?<br>
                        <br>
                        I think conceptually it was as simple as a
                        non-signed JWT containing iss and token fields.
                        Obviously, the rest of JOSE could be applied for
                        signed or encrypted tokens.<br>
                        <br>
                        Thanks,<br>
                        George<br>
                        <br>
                      </font>
                      <div>On 3/12/14 9:02 PM, n-sakimura wrote:<br>
                      </div>
                      <blockquote type="cite">Let's just write up
                        requirements on the WG wiki (@bitbucket). <br>
                        Once we agree on the requirements, it should be
                        straight forward to turn it into a spec. <br>
                        <br>
                        On the side note, perhaps it is actually for
                        OAuth WG, but it would be nice to spec out the
                        structured (access) token. it could be pseudo
                        opaque as well as long as you can find the
                        authorization server from the token but we at
                        least need to be able to find out the iss. <br>
                        <br>
                        Nat <br>
                        <br>
                        (2014/03/13 3:58), John Bradley wrote: <br>
                        <blockquote type="cite">We have discussed
                          creating a backchannel push method for the IdP
                          to notify the RP. <br>
                          <br>
                          So far noting is written up.  I have a bad
                          feeling that it might be me that needs to
                          create the first draft. <br>
                          <br>
                          John B. <br>
                          <br>
                          On Mar 12, 2014, at 3:54 PM, Pedro Felix <a
                            moz-do-not-send="true"
                            href="mailto:pmhsfelix@gmail.com"
                            target="_blank"><pmhsfelix@gmail.com></a>
                          wrote: <br>
                          <br>
                          <blockquote type="cite">Hi, <br>
                            <br>
                            I've a scenario where a OIDC OP is acting as
                            a bridge between upstream IdPs using
                            non-OIDC protocols (e.g Shibboleth) and
                            downstream RPs using OIDC. <br>
                            In this scenario I have the following
                            requirements <br>
                               1) The upstream IdP notifies the OP of a
                            session termination via back-channel <br>
                               2) The OP propagate this cleanup
                            notification to the downstream RPs, also via
                            back-channel (a back-channel to
                            front-channel is not possible) <br>
                            <br>
                            Unfortunately, the OIDC session management
                            spec does not provide any way to perform
                            this back-channel cleanup, however I
                            remember reading some meeting notes about
                            this possibility. <br>
                            <br>
                            Is there anything that can be shared? I
                            would like to align our solution with what
                            is being developed by this working group. <br>
                            <br>
                            Thanks <br>
                            Pedro <br>
                            _______________________________________________
                            <br>
                            Openid-specs-ab mailing list <br>
                            <a moz-do-not-send="true"
                              href="mailto:Openid-specs-ab@lists.openid.net"
                              target="_blank">Openid-specs-ab@lists.openid.net</a>
                            <br>
                            <a moz-do-not-send="true"
                              href="http://lists.openid.net/mailman/listinfo/openid-specs-ab"
                              target="_blank">http://lists.openid.net/mailman/listinfo/openid-specs-ab</a>
                            <br>
                          </blockquote>
                          <br>
                          <br>
                          <br>
                          _______________________________________________
                          <br>
                          Openid-specs-ab mailing list <br>
                          <a moz-do-not-send="true"
                            href="mailto:Openid-specs-ab@lists.openid.net"
                            target="_blank">Openid-specs-ab@lists.openid.net</a>
                          <br>
                          <a moz-do-not-send="true"
                            href="http://lists.openid.net/mailman/listinfo/openid-specs-ab"
                            target="_blank">http://lists.openid.net/mailman/listinfo/openid-specs-ab</a>
                          <br>
                          <br>
                        </blockquote>
                        <br>
                        <br>
                      </blockquote>
                      <br>
                      <div>-- <br>
                        <a moz-do-not-send="true"
                          href="http://connect.me/gffletch" title="View
                          full card on Connect.Me" target="_blank"><img
                            src="cid:part7.02070104.05070005@aol.com"
                            alt="George Fletcher" width="359"
                            height="113"></a></div>
                      <br>
                      <fieldset></fieldset>
                      <br>
                      <pre>_______________________________________________
Openid-specs-ab mailing list
<a moz-do-not-send="true" href="mailto:Openid-specs-ab@lists.openid.net" target="_blank">Openid-specs-ab@lists.openid.net</a>
<a moz-do-not-send="true" href="http://lists.openid.net/mailman/listinfo/openid-specs-ab" target="_blank">http://lists.openid.net/mailman/listinfo/openid-specs-ab</a>
</pre>
                    </blockquote>
                    <br>
                  </div>
                </div>
              </div>
              <br>
              _______________________________________________<br>
              Openid-specs-ab mailing list<br>
              <a moz-do-not-send="true"
                href="mailto:Openid-specs-ab@lists.openid.net">Openid-specs-ab@lists.openid.net</a><br>
              <a moz-do-not-send="true"
                href="http://lists.openid.net/mailman/listinfo/openid-specs-ab"
                target="_blank">http://lists.openid.net/mailman/listinfo/openid-specs-ab</a><br>
              <br>
            </blockquote>
          </div>
          <br>
        </div>
      </div>
    </blockquote>
    <br>
    <div class="moz-signature">-- <br>
      <a href="http://connect.me/gffletch" title="View full card on
        Connect.Me"><img src="cid:part13.07010303.04030702@aol.com"
          alt="George Fletcher" width="359" height="113"></a></div>
  </body>
</html>